Traditionally, the ability of experimentalists to perturb a system has been limited to rigid pre-programmed patterns or flexible operator-controlled protocols constrained by human reflexes. In contrast, "real-time control" allows the researcher to utilize dynamic manipulations according to automated analysis of instantaneous system measurements, thereby enabling entirely new biological experimentation paradigms in fields ranging from molecular biophysics to neuroscience. Unfortunately, real-time control is not possible with standard computer operating systems, and commercial real-time systems are costly and often tailored for industrial non-biological applications. The work proposed here is aimed at bypassing these problems by developing a freely available, powerful, and flexible open-source system for real-time control of biological experiments. The specific aims of this project are: 1. To unify three complementary real-time control systems. The investigator and co-investigators have been independently developing real-time control systems. Each system has particular strengths that complement the weaknesses of the others. We will merge these systems to form a powerful unified system with greatly expanded functionality. 2. To extend functionality to non- Linux platforms. Because our goal is to enable real-time biological experimentation, rather than to convert users to Linux, we will develop methods by which Windows and Apple users can run our system without having to abandon their preferred platforms. 3. To develop performance-monitoring metrics and tools. To aid experimentalists in designing an experiment application, tools for monitoring system performance will be developed and performance metrics will be established. 4. To create an extensive real-time experiment application library. We will develop a library of both general-purpose and application-specific experiment-control "plugins". Users will be able to choose from these options and/or customize them to meet their particular experiment demands. 5. To develop a visual-design tool for plugin development. We will develop a visual-design tool with which users will be able to develop plugins without manually editing source code. By eliminating the need for programming, this tool will greatly increase the number of users who will be able to develop experiment-control applications. 6. To optimize ease of use. To minimize the barrier of entry for beginners and maximize the productivity of experienced users, we will emphasize user-friendliness and develop comprehensive documentation. [unreadable] [unreadable]